The point is that GXT isn't a JS Wrapper. It's entirely written in
Java.

GWT-EXT and SmartGWT are Javascript Wrappers.

Tercio


On Jan 18, 2:51 pm, Arthur Kalmenson <[email protected]> wrote:
> JS wrappers won't benefit from the GWT compiler, so they would
> theoretically be slower. Have you tried without GXT? Is it much
> faster? We don't use GXT here and have no issues with dev mode
> startup, it's very fast.
>
> --
> Arthur Kalmenson
>
>
>
> On Tue, Jan 12, 2010 at 10:06 PM, Tercio <[email protected]> wrote:
> > Hi!
>
> > I'm on a Mac 10.6.2 an I have the same issue here. My test as on
> > WebKit Version 4.0.4 (6531.21.10, r52686).
>
> > Using GWT 2.0 and GXT 2.1.0.
>
> > I think that the problem is the combination of GXT and GWT. I tested
> > my application without Hibernate and it is a little faster, but still
> > very slow.
>
> > Maybe as GXT is pure JAVA it needs some parse from the GWT, other
> > solutions, like SmartGWT that is just a wrapper it may be faster, I
> > don't know.
>
> > If I compile my project and run it as javascript it gets stunning
> > fast!
>
> > Regards,
>
> > On Jan 12, 8:22 pm, Chris Lowe <[email protected]> wrote:
> >> Hi Tim,
>
> >> It's still conceivable for a circular reference (or at least massively
> >> repeated objects) to be at play here.  Your response size is 165739 bytes
> >> *compressed* size - many identical objects could compress to something
> >> relatively small and their expansion could cause issues.  Also with
> >> Hibernate you most likely won't see a lot of database traffic from repeated
> >> objects as subsequent fetch requests will hit the session's level 1 cache.
>
> >> A quicker test could be to temporarily disable gzip encoding in your
> >> browser, for example here are instructions for FireFox:
>
> >>http://kb.mozillazine.org/Network.http.accept-encoding
>
> >> With gzip disabled, what size response does Jetty report?
>
> >> BTW - which browser are you using?
>
> >> Cheers,
>
> >> Chris.
>
> >> <http://kb.mozillazine.org/Network.http.accept-encoding>
>
> >> 2010/1/12 Tim Mattison <[email protected]>
>
> >> > That's what I thought originally but I can see that it's only pulling 
> >> > back
> >> > 165739 bytes from the RPC.  When I don't return anything it turns out 
> >> > that
> >> > it runs quickly so it obviously must be related to that.
>
> >> > I'm not using DTOs but where is it trying to fetch the data from if it's
> >> > not getting pulled back from the POST?  I don't see any traffic to the
> >> > database server so it must be doing something weird locally (like a 
> >> > circular
> >> > reference as you suggested).  Shouldn't it give up on circular 
> >> > references a
> >> > bit faster than that if that's the case?
>
> >> > Tim
>
> >> > On Tue, Jan 12, 2010 at 3:56 PM, Chris Lowe 
> >> > <[email protected]>wrote:
>
> >> >> Tim,
>
> >> >> Are you filtering your Hibernate objects or translating them to DTOs (to
> >> >> remove dynamic proxies etc) before serialising them?
>
> >> >> If the answer is no to the above, then you might be falling foul to
> >> >> circular references or Hibernate fetching more data than you expect.  
> >> >> As an
> >> >> experiment, is it possible to try hardcoding some of your objects with a
> >> >> minimal data set and see how your app performs?
>
> >> >> Cheers,
>
> >> >> Chris.
>
> >> >> 2010/1/12 Tim Mattison <[email protected]>
>
> >> >>> I changed my debug level from "Info" to "Debug" and got lots of
> >> >>> additional output but nothing that looked like it was the culprit.  My
> >> >>> application runs like this:
>
> >> >>> 1) onModuleLoad is called, builds the UI, and fires off a GWT-RPC call
>
> >> >>> 2) The server receives the GWT-RPC call, connects to a Hibernate
> >> >>> database, pulls some data (~150K) and sends it to the client
>
> >> >>> 3) The client receives the response and populates a FlexTable with the
> >> >>> data
>
> >> >>> Between 2 and 3 is where the storm of traffic occurs.  With the new 
> >> >>> debug
> >> >>> level I don't really get much more insight since I see that Jetty has 
> >> >>> sent
> >> >>> the response to the browser and that's it.  I have breakpoints set on 
> >> >>> my
> >> >>> GWT-RPC callback's onFailure and onSuccess method and it doesn't get to
> >> >>> either of those branches until minutes later.  Is there somewhere else 
> >> >>> I can
> >> >>> look or something else I can try?
>
> >> >>> The last message in the log before the storm:
>
> >> >>> 200 - POST /app/service (127.0.0.1) 165739 bytes
>
> >> >>>    Request headers
>
> >> >>>       Host: localhost:8888
>
> >> >>>       User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; 
> >> >>> en-US;
> >> >>> rv:1.9.1.7) Gecko/20091221 Firefox/3.5.7
>
> >> >>>       Accept:
> >> >>> text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
>
> >> >>>       Accept-Language: en-us,en;q=0.5
>
> >> >>>       Accept-Encoding: gzip,deflate
>
> >> >>>       Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
>
> >> >>>       Keep-Alive: 300
>
> >> >>>       Connection: keep-alive
>
> >> >>>       Cache-Control: no-cache
>
> >> >>>       Referer:http://localhost:8888/app/hosted.html?app
>
> >> >>>       X-GWT-Permutation: HostedMode
>
> >> >>>       X-GWT-Module-Base:http://localhost:8888/app/
>
> >> >>>       Content-Type: text/x-gwt-rpc; charset=utf-8
>
> >> >>>       Content-Length: 175
>
> >> >>>       Pragma: no-cache
>
> >> >>>    Response headers
>
> >> >>>       Content-Encoding: gzip
>
> >> >>>       Content-Length: 165739
>
> >> >>>       Content-Type: application/json; charset=utf-8
>
> >> >>>       Content-Disposition: attachment
>
> >> >>> The first message after it hits onSuccess and then keeps going at a
> >> >>> normal speed:
>
> >> >>> 304 - GET /app/gwt/standard/images/hborder.png (127.0.0.1)
>
> >> >>>    Request headers
>
> >> >>>       Host: localhost:8888
>
> >> >>>       User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; 
> >> >>> en-US;
> >> >>> rv:1.9.1.7) Gecko/20091221 Firefox/3.5.7
>
> >> >>>       Accept: image/png,image/*;q=0.8,*/*;q=0.5
>
> >> >>>       Accept-Language: en-us,en;q=0.5
>
> >> >>>       Accept-Encoding: gzip,deflate
>
> >> >>>       Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
>
> >> >>>       Keep-Alive: 300
>
> >> >>>       Connection: keep-alive
>
> >> >>>       Referer:http://localhost:8888/app/gwt/standard/standard.css
>
> >> >>>       If-Modified-Since: Tue, 03 Nov 2009 15:44:06 GMT
>
> >> >>>       Cache-Control: max-age=0
>
> >> >>>    Response headers
>
> >> >>> Any help would be great.
>
> >> >>> Tim
>
> >> >>> On Tue, Jan 12, 2010 at 3:04 PM, Chris Ramsdale 
> >> >>> <[email protected]>wrote:
>
> >> >>>> Although this smells of a network configuration issue, one suggestion
> >> >>>> you could try is to set the log level to Debug or lower.
>
> >> >>>> Debug->Debug Configurations->GWT->Log level.
>
> >> >>>> Try that, and let us know if anything suspect is output.
>
> >> >>>> - Chris
>
> >> >>>> On Mon, Jan 11, 2010 at 11:56 AM, timmattison 
> >> >>>> <[email protected]>wrote:
>
> >> >>>>> I just started using OOPHM on my Mac (10.6.2) and it is very, very
> >> >>>>> slow.  I've tried all of the recommendations about changing the URL 
> >> >>>>> to
> >> >>>>> include only "localhost" or "127.0.0.1" but I still have to wait
> >> >>>>> nearly three minutes for my application to start.
>
> >> >>>>> The program I'm writing is currently very small and only consists of
> >> >>>>> less than 200 lines of code.  It does import a JAR that contains
> >> >>>>> definitions of a lot of objects and has some dependencies (Gilead,
> >> >>>>> Hibernate, GXT) for the server side components but right now I'm just
> >> >>>>> using basic GWT components.  Does the size of the dependencies and
> >> >>>>> included JARs matter?
>
> >> >>>>> I ask because I notice that as soon as I start the application the
> >> >>>>> traffic on port 9997 to and from my loopback interface is pegged at
> >> >>>>> 1.5MB/sec in each direction for the entire three minutes the
> >> >>>>> application is starting up.  I stepped through my code with a 
> >> >>>>> debugger
> >> >>>>> and the client side code gets set up, runs, then there's a three
> >> >>>>> minute pause where all of this data goes back and forth, and then the
> >> >>>>> server-side code runs.  The client and server side code takes less
> >> >>>>> than 1 second to finish so I don't think it's a bug in my code.
>
> >> >>>>> I tried to capture the traffic in Wireshark to figure out what is
> >> >>>>> getting sent but it looks like all of the packets are very small (~56
> >> >>>>> bytes) and trying to capture the whole session causes Wireshark to
> >> >>>>> crash.
>
> >> >>>>> Is anyone else seeing this loopback traffic problem?  I assumed maybe
> >> >>>>> the debugger is communicating my dependencies to the OOPHM plugin but
> >> >>>>> my dependencies are nowhere near this large.
>
> >> >>>>> What other information can I provide to help this get debugged?
>
> >> >>>>> Thanks,
> >> >>>>> Tim Mattison
>
> >> >>>>> --
> >> >>>>> You received this message because you are subscribed to the Google
> >> >>>>> Groups "Google Web Toolkit" group.
> >> >>>>> To post to this group, send email to
> >> >>>>> [email protected].
> >> >>>>> To unsubscribe from this group, send email to
> >> >>>>> [email protected]<google-web-toolkit%2Bunsubs
> >> >>>>>  [email protected]>
> >> >>>>> .
> >> >>>>> For more options, visit this group at
> >> >>>>>http://groups.google.com/group/google-web-toolkit?hl=en.
>
> >> >>>> --
> >> >>>> You received this message because you are subscribed to the Google
> >> >>>> Groups "Google Web Toolkit" group.
> >> >>>> To post to this group, send email to
> >> >>>> [email protected].
> >> >>>> To unsubscribe from this group, send email to
> >> >>>> [email protected]<google-web-toolkit%2Bunsubs
> >> >>>>  [email protected]>
> >> >>>> .
> >> >>>> For more options, visit this group at
> >> >>>>http://groups.google.com/group/google-web-toolkit?hl=en.
>
> >> >>> --
> >> >>> You received this message because you are subscribed to the Google 
> >> >>> Groups
> >> >>> "Google Web Toolkit" group.
> >> >>> To post to this group, send email to 
> >> >>> [email protected]
> >> >>> .
> >> >>> To unsubscribe from this group, send email to
> >> >>> [email protected]<google-web-toolkit%2Bunsubs...
>
> read more »
-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/google-web-toolkit?hl=en.


Reply via email to